Mathematical modelling to project the impact of public health interventions targeted to individuals previously treated for tuberculosis on the trajectory of the tuberculosis epidemic and the emergence of drug-resistant diseae in Southern Africa
Epidemiology and Medical Biometry/Statistics
Final Report Abstract
Recurrent tuberculosis (TB), defined as TB that re-occurs after a patient had been considered cured, constitutes an important challenge to TB control. In recent years, high rates of recurrent TB after curative treatment have been reported from several TB high-incidence settings in Southern Africa. Both endogenous reactivation and exogenous reinfection with Mycobacterium tuberculosis (M.tb) contribute to recurrence in these settings. Rates of recurrent TB appear to be high among both HIV infected and uninfected people and exceed the rate of new TB several-fold. The extent to which individuals previously treated for TB contribute to overall M.tb transmission and the disease burden in TB high-incidence populations is currently not known. In this research project, we used traditional epidemiological analysis methods and mathematical modeling to quantify the burden of prevalent TB among people with a history of previous TB treatment in settings of high TB incidence, and to project the population-level impact of control interventions targeted toward this high-risk group. Analysis of TB prevalence survey data from 24 African high-burden communities showed that adults previously treated for TB represent a variably large fraction of the adult population, which is most sizeable (up to 15%) in communities with the highest TB burden. We found a high prevalence of undetected TB in this subgroup, unexplained by HIV infection, and on average nearly twice as high as that among people without previous TB treatment. Previously treated TB accounted for a considerable proportion of the overall prevalent TB burden (>20% in nine communities). We further developed a transmission-dynamic compartmental mathematical model of the TB epidemic to project the population-level impact of (1) annual targeted active TB case finding (TACF) among individuals who had previously completed TB treatment and (2) TACF combined with lifelong secondary isoniazid preventive therapy (2°IPT). Using a Bayesian posterior estimation approach, we calibrated our model to epidemiological, demographic, and TB control data from a high-incidence setting in suburban Cape Town, South Africa. The model projected that the TB epidemic in this setting will remain in slow decline for at least the next decade. TACF and 2°IPT among people who had previously completed TB treatment would greatly accelerate these declines. We projected that the implementation of TACF alone would avert 14% of incident TB cases and 21% of TB deaths between 2016 and 2025; the implementation of TACF in combination with 2°IPT would avert 40% of incident TB cases and 41% of TB deaths in the 10-year period. The findings from this research project reveal that people with a history of previously treated TB represent a large population subgroup that contributes considerably to the prevalent TB burden in high-burden communities in Southern Africa. Mathematical modeling suggests considerable public health potential of control interventions targeted toward this high-risk group. Our study constitutes a first step towards better understanding of the role of former TB patients as a risk group for TB in high-incidence settings. More research is needed to understand if targeting TB control interventions toward individuals previously treated for TB can be feasible and effective, and whether these additional control efforts could be cost-effective for TB control in populations most affected by the disease.
Publications
- High burden of prevalent tuberculosis among previously treated people in 24 African communities suggests potential for targeted control interventions. Eur Respir J. 2016 Jul 7. pii: ERJ- 00716-2016
FM Marx, S Floyd, H Ayles, P Godfrey-Faussett, N Beyers, T Cohen
(See online at https://dx.doi.org/10.1183/13993003.00716-2016) - Impact of tuberculosis control interventions targeted to previously treated people in a high-burden setting. (Scientific Abstract) Int J Tuberc Lung Dis 2016: 20 (11 Suppl 1): S1–S521
FM Marx, R Yaesoubi, NA Menzies, JA Salomon, N Beyers, T Cohen
